Research on Load Balancing Method of Object Storage System Based on Data Heat Prediction and Migration

  • Hao LiEmail author
  • Yi Chai
  • Ke Zhang
  • Qiulin Dan
Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 528)


Since most existing object storage systems balance the load by scheduling task requests, there is no concern about the uneven load generated by random access to the data, resulting in a large number of requests to be concentrated on a small number of servers. Based on Weighted Least-Connection (WLC) and Consistent Hashing, a load balancing algorithm for Heat Prediction and Migration (HPM) is proposed. By refining the number of connections to the object layer and comprehensively considers the number of connections, the difference of heterogeneous nodes and objects as the object heat, and predicts the heat to guide the scheduling of objects. Finally, the effectiveness of the algorithm is verified by simulation.


Object storage Load balancing Heat prediction and migration 


  1. 1.
    M. Mesnier, G.R. Ganger, E. Riedel, Object-based storage. Commun. Mag. IEEE 41(8), 84–90 (2003)CrossRefGoogle Scholar
  2. 2.
    M. Mesnier, G. Ganger, E. Riedel, Object-based storage: pushing more functionality into storage. Potent. IEEE 24(2), 31–34 (2005)CrossRefGoogle Scholar
  3. 3.
    L. Yan, D. Feng, L. Zeng et al., Dynamic load balancing algorithm based on object storage System. Comput. Sci. 33(5), 88–91 (2006)Google Scholar
  4. 4.
    H. Yang, B. Lin, Research on consistent hash algorithm in distributed storage system. Comput. Knowl. Technol. 07(22), 5295–5296 (2011)Google Scholar
  5. 5.
    Y. Luo, X. Li, R. Sun, A review of load balancing algorithms. Library Informat. Sci. 18(23), 134–136 (2008)Google Scholar
  6. 6.
    Z. Gongye, W. Lei, J. Chen, Hotspot data balancing strategy based on object storage system. J. Huazhong Univ. Sci. Technol. (Nature Sci.), 2007, 35(12): 28–31Google Scholar
  7. 7.
    D. Zhao, Research and Implementation of Read Load Balancing in Swift (Southeast University, 2016), 8Google Scholar
  8. 8.
    Z. Wang, H. Chen, Y. Fu et al., Workload balancing and adaptive resource management for the swift storage system on cloud. Fut. Generat. Comput. Syst. 51, 120–131 (2015)CrossRefGoogle Scholar
  9. 9.
    X. Min, M. Li, J. Zheng et al., Swift load balancing algorithm based on OpenStack. J. Comput. Syst. 1, 127–131 (2018)Google Scholar
  10. 10.
    R. Lee, B. Jeng, Load-balancing tactics in cloud, in International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (IEEE, 2011), pp. 447–454Google Scholar
  11. 11.
    Z. Zhang, X. Zhang, A load balancing mechanism based on ant colony and complex network theory in open cloud computing federation, in The, International Conference on Industrial Mechatronics and Automation (2010) 240–243Google Scholar
  12. 12.
    A. Graves, Long short-term memory supervised sequence labelling with recurrent neural networks (Springer, Berlin Heidelberg, 2012), pp. 1735–1780CrossRefGoogle Scholar
  13. 13.
    J. Yingmin, Robust control with decoupling performance for steering and traction of 4WS vehicles under velocity-varying motion. IEEE Trans. Control Syst. Technol. 8(3), 554–569 (2000)CrossRefGoogle Scholar
  14. 14.
    Y. Jia, Alternative proofs for improved LMI representations for the analysis and the design of continuous-time systems with polytopic type uncertainty: a predictive approach, IEEE Trans. Automat. Control 48(2003), 8, 1413–1416Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2019

Authors and Affiliations

  1. 1.State Key Laboratory of Power Transmission Equipment and System Security and New Technology, College of AutomationChongqing UniversityChongqingChina

Personalised recommendations